Publication Date
IT infrastructures on the way to a new balance
The adoption of cloud computing has seen explosive growth over the past decade, enough to become a commodity and, at times, be regarded as a solution for all IT challenges.
However, since 2023 there has been a trend toward the partial recentralization of certain systems. This is not a rejection of the cloud, but rather a desire by organizations to regain control over specific elements of their IT infrastructure. Regulatory obligations, such as Law 25, require stricter supervision regarding the location and processing of data.
Many companies have discovered that the cloud, without rigorous governance, can become more expensive than anticipated. The “pay-as-you-go” model is very attractive at first, but can lead to surprises if poorly managed. In many cases, a lack of understanding of cost-reduction mechanisms explains budget overruns. To this, one must add the shortage of certain specialists capable of effectively managing these infrastructures. Managing cloud environments requires specialized skills and sound governance.
For many organizations, the challenge is to determine the right balance between cloud, on-premises infrastructure, or a hybrid approach, depending on their specific needs and constraints. The choice of platform and services should be based on the elements that bring the greatest value to the organization and its clients. To make this decision, what factors should be taken into account?
A decision that starts with your data
Before thinking about technology, consider your data.
- Where is your data generated?
- Where does it need to be processed?
- How quickly do you need to access it?
- What is the level of criticality for each?
A good indicator is what we call Data Gravity; a metaphor borrowed from astronomy to explain the pull that data exerts on the systems that use it. For example, if your main database is in the cloud, it often makes sense to also host your applications there, to avoid costly or slow roundtrips between your systems and your data. Conversely, if your industrial equipment, for example, generates real-time data on-site, it is in your best interest to process it locally, near its source, to avoid delays and the risk of disconnection.
Understanding data gravity helps you choose the optimal location for your IT systems. It’s not just a matter of technology, but one of operational logic and efficiency.
On which planet should you land?
Now that you have a clearer understanding of your different types of data and the services that support them, it’s time to choose the right “landing ground(s)” for the systems that use them. You need to determine the environment that will allow you to extract maximum value. In other words, where to host your applications, services, and IT resources so they are efficient, secure, and sustainable, all while respecting the concept of data gravity that we’ve just explained.
There are three major options, each with its strengths, limitations, and preferred areas of application. None is inherently better than the other, as your choices will also be influenced by your business context, internal resources, and strategic priorities.
1. On-premises infrastructure
This is the classic model. Your servers and equipment are installed on your premises or in a data center that you manage. This means more control, but also more responsibilities: updates, security, backups, and so on.
- Advantages: Total control, data proximity, compliance.
- Disadvantages: High initial investment, ongoing maintenance, risk of obsolescence. Emerging technologies often take longer to become available on site and require advanced expertise. Redundancy and disaster recovery plans must also be provided, which increases costs.
Ideal for environments requiring strict data control and low latency, such as industrial settings or infrastructures supporting critical data.
2. The cloud (cloud computing)
There is the public cloud (Azure, AWS, Google Cloud), private, or even shared. In this model, your data and applications are hosted remotely and accessed via the Internet. This reduces hardware investments but implies a dependency on an external provider.
- Advantages: Flexibility, rapid deployment, scalable resources on demand, simplified access to innovative technologies (AI, advanced analytics, etc.), and cost models tailored to consumption.
- Disadvantages: Dependence on a provider and the need for increased governance to manage security and control costs.
Particularly suited for growing or transforming businesses that need to rapidly deploy remote-accessible services and manage variable resource requirements. This model also enables easy access to modern or specialized services that require advanced expertise, or that would be difficult to obtain and/or afford on-premises.
3. Hybrid infrastructure
Historically, hybrid infrastructure meant combining on-premises and cloud solutions—the best of both worlds. With this approach, you can keep certain critical applications on site and migrate other services to the cloud. This allows you to adapt to needs, regulatory constraints, and internal preferences, while benefiting from modern or emerging services that are often available only in the cloud, or at much more competitive costs than on site.
- Advantages: Adaptability, gradual transition, increased resilience.
- Disadvantages: Management complexity, need for effective orchestration of data flows, and sometimes reliance on modern or highly specialized services that are not readily available or affordable on site.
The hybrid model remains a preferred path for balancing compliance, criticality, and agility.
The advent of sovereign options in the cloud enhances its appeal and demonstrates that security and compliance can now be achieved just as effectively in a hybrid setting as in a fully cloud-based environment.
Today, we are fully entering the era of gravity, where data, users, and regulatory requirements dictate where services must reside. This strategic approach becomes even more crucial with the emergence of new sovereign solutions, such as those announced by Microsoft in June 2025 for Europe.
With this announcement, Microsoft renews its commitment with Microsoft Sovereign Cloud, now available in three complementary offerings:
- Sovereign Public Cloud: data remains in the country, under local legislation, accessible only by local personnel, with encryption managed by clients.
- Sovereign Private Cloud: allows local, isolated or private Azure environments to host Exchange, SharePoint or Microsoft 365, with full control over security and continuity.
- National Partners Clouds: operated in partnership with national providers to offer solutions that meet local legal requirements.
This European wave marks a strategic turning point that will eventually impact Canada. In this new context, hybrid infrastructures are no longer just modular; they become environments with variable gravity, able to adapt the location, protection, and sovereignty of data at will.
In a world where gravity shifts according to sectors (health, finance, governance…) and geopolitical risks, choosing the right gravity becomes a lever for trust and competitiveness. It is no longer just a technical architecture—it’s a strategic stance.
Choosing your IT infrastructure today is much more than a technological decision—it’s a foundational strategic act. In the age of Gravity-first, your choices must align with where your data lives, evolves, and generates value. Above all, they must support your business priorities: agility, flexibility, speed to market, making data accessible to the right people at the right time, and sustainable competitiveness with your clients.
There is no one-size-fits-all solution, but there is an optimal setup that respects your constraints, ambitions, and operational reality.
Key Takeaways
By asking the right questions now, you’ll be able to build an infrastructure that doesn’t lock you in, but instead adapts as you grow. What you put in place today must be able to follow you tomorrow, no matter which direction your gravity takes.
